Given that we've since had a change of Prime Minister, various attempted coups against Gordon Brown and been teased about early polls that never came to pass, it seems like an eternity since the last General Election. My regret about this isn't just political. Of all the various sports and markets that I make my living betting on, nothing even comes close to an election.
Taking risks on political betting is nothing like sport. As with other novelty markets, (i.e. Big Brother, X Factor), there are simply fewer random effects to account for. We can all confidently assert that Chelsea will beat Hull City, but there are no secrets and the odds will reflect their overwhelming superiority. However, once every so often, the outsiders will play a blinder, Chelsea might hit the woodwork six times, and punters who've shelled out a fortune at miniscule odds about the favourites will learn a lesson they'll never forget.
That crucial element of 'luck' is removed when betting on politics. If I predict, for example, the Lib Dems to win City of Durham by 2,000 votes plus, I can only lose if I'm badly wrong. 2,000 people are not going to all be impeded on their way to the polling booths in Durham.
Five years ago, such confidence proved justified . After analysing every seat, my prediction was Labour 359, Conservative 191, Lib Dem 67. The actual shares were 356, 196 and 62, ensuring nearly all the bets on seat totals and size of majority were landed.
In fairness, politics was much more predictable back then. During the long economic boom, the opinion polls barely shifted, showing a clear Labour lead with the Tories stuck in the low 30s. The only really significant swing in 2005 was away from Labour towards the Lib Dems, primarily thanks to their opposition to the Iraq war.
That predictability is long gone. The last Parliament quickly saw the Tories, under a more personable, swing-voter friendly leader in David Cameron, open up big opinion poll leads. As the Lib Dems twice switched leaders and all but disappeared from mainstream media coverage, Cameron was able to position himself as the default alternative to an unpopular government.
However even at their high mid-term point, the Tories never truly captured public sentiment. Indeed, a clear sign of the electorate's scepticism about Cameron's Tories was evident when Gordon Brown suddenly re-established a lead after replacing Tony Blair as PM. Had he held that infamous election that never was in late 2007, Labour would have been heavily odds-on to win the most seats. Once Brown famously 'bottled' that early election, and the economy turned very sour, the opinion polls swung further towards the Tories than for at least three decades.
During the last three years, the long-running ICM/Guardian poll series has shown variations between a Labour lead of six points, and a Tory lead of twenty. Now, since the dramatic advance of the Lib Dems during this campaign, the latest of these 'gold standard' polls shows all three parties separated by just seven points. The electorate has never been so volatile, reflecting a wider disenchantment with both the process and political class in general. It all makes for the most exciting election in living memory.
How that public mood plays out in terms of seat totals depends on a variety of regional nuances and tactical factors. We should treat predictions from the TV news with great caution, even on the night itself. Too often, their projections rely on simplistic assumptions about national swing based on the headline voting figures, implying that there is a uniform effect across the whole country and that seats will be distributed accordingly. Two examples involving the Lib Dems illustrate why this isn't the case.
In 2005, when the BBC coverage started, showing exit polls that turned out to be broadly accurate, their estimate was for the Lib Dems to get around 50 seats. Consequently, the in-running markets on Betfair reflected those predictions and some astonishing value appeared either to lay them getting below 50, or to back them to get 60+. Remember, this is after all the votes had been cast. The result was 62.
Equally, as their polls have improved massively in recent weeks, we've heard some wild projections about their seat tally. Even though these projections already recognise that the voting system is hopelessly biased against the Lib Dems, they seem unable to reflect the fact that Lib Dem support is particularly concentrated in certain regions, amongst specific type of voter. They don't have the resources to run strong local campaigns beyond the first 120-odd seats they either hold or are realistically targetting. As a Lib Dem supporter, I take no pleasure in saying this, but even on a brilliant night where they won the popular vote, they'll struggle to get many more than 100.
The converse effect is that the Lib Dems should be very reliable in the seats that they are targetting, and this is where most of the best value in the constituency betting lies. Before I get to each party's best targets, here's my prediction of the overall results.
POLLS AND SEAT PREDICTIONS
I've deliberately tried to leave this to the last moment, so as not to be wrong-footed by debate performances or events on the campaign trail. Given that the latest polls were taken since 'Bigotgate' and the final TV debate, I doubt we'll see much movement before Thursday.
CONSERVATIVES 36% (313 SEATS)
LIBERAL DEMOCRATS 29% (100 SEATS)
LABOUR 25% (204 SEATS)
OTHERS 10% (33 SEATS)
Note these predictions are optimistic from a Tory perspective and negative for Labour, compared to current estimates, (yesterday's Yougov/Sun daily tracker had scores of CON 34/LD 28/LAB 28, for instance). Two reasons lead me to think the Tories will perform slightly better than that poll suggests. Firstly, their support is likelier to be registered to vote and more motivated to turn out. In contrast, getting their increasingly distant core vote to the polling booths has been a long-term problem for Labour, while this extra Lib Dem support stems mainly from young people, amongst whom turnout is lower. Secondly, opinion polls have repeatedly over-estimated Labour in previous decades, partly for those reasons of turnout.
Even on these optimistic figures, there's no real evidence of a seismic shift in public opinion towards the Tories. Considering they scored 32.3% in 2005, under an unpopular leader in Michael Howard, before the recession, there don't appear to be many new Tory voters out there. Those that have switched are probably pre-1997 supporters, rather than new, previously anti-Tory segments of the population. Nevertheless, the swing will be enough to ensure they comfortably win the most seats, and to make David Cameron PM. My prediction leaves them 13 short of an overall majority, but enough to form a workable minority government without forming a coalition.
They can win so many seats despite such a small rise in their share of the vote, because the Labour vote is collapsing. Clearly, there is a groundswell of opinion that its time for a change. Most switchers may have decided that the change should be in a Lib Dem, rather than Conservative, direction, yet the consequence will be to make the Tories' task easier in dozens of critical Labour-held marginals. To take the fairly typical example of a constituency where the 2005 result was Lab 42, Con 30, LD 18. If the Lab vote drops by 10%, 7% of which go Lib Dems and 3% to the Conservatives, Cameron's party would win on 33.
Basically, the Tories will take the world of beating in any Labour-held seat where they scored 30% last time. That is enough to become the largest party with ease, but in order to gain a sizeable majority, they'll need to start winning over people who've never previously voted Tory. That was possible when the election looked a 2-horse race, less so now Clegg has stolen the 'change agenda'.
Those seat total predictions are based on analysis of every single constituency stretching back three years. Here's how it breaks down.
To start, I took the 200 likeliest Tory gains from Labour, including some highly unlikely places. 77 are confidently fancied to go Tory, 44 very much so. In itself, that ensures the Tories win the most seats. 52 are fancied to stay Labour. A further 25 Con/Lab marginals are still deemed too close to call, though for the sake of the overall prediction I have allocated 15 to the Tories, 10 to Labour. Then there's the 46 of these seats where the Lib Dems could play a significant hand. Out of these, I'm only predicting the Tories will win 12. So here, the total is 104 Tory gains from Labour.
A couple of months ago, that kind of result looked like it could be enough to secure a small Tory majority, because they looked certain to make substantial gains from the Lib Dems. Six months ago, I'd have predicted 20+. Suddenly, due to the outbreak of Cleggmania following the Lib Dem leader's starring performance in the TV debates, the reverse effect seems vaguely possible. In my view, several seats will change hands in either direction, but there will be little or no net change. For the sake of the calculations, I'm predicting no net change. Against Labour, my estimate is that the Lib Dems will gain 38.
BEST CONSERVATIVE BETS
While there is obviously a margin of error with these predictions, given the unusually high number of 'too close to calls' and unforeseen local factors, I'll be very surprised if they're more than 20 off the mark. In previous elections, picking the correct band in the seat totals markets has been pretty easy and I think it is once again. In the Tories' case, the best bet is 5/2 about the 300 - 324 band, along with a saver on 325 - 349 at 3/1. I see very little risk in such a strategy.
Of course, given that they're heavily odds-on to win the most seats, they are a similarly prohibitive price for their likeliest gains. Brave punters might be best served focussing on a couple of seats where the Labour lead may have been previously exaggerated due to a popular long-term MP, KEIGHLEY and CREWE & NANTWICH . In the former, Anne Cryer is standing down, making it a prime target to return to pre-1997 status. The latter was famously won in a 2008 by-election, at what may prove to be the height of Cameron's popularity. The scale of the victory was stunning, even if the result wasn't quite the surprise it was made out to be, as the former Labour incumbent, the late Gwyneth Dunwoody, almost certainly enjoyed a vast personal vote.
Three Labour-held seats in Yorkshire, where the Tories are outsiders, look decent value. The relevant swing appears to be bigger in this heavily targetted region. While the Ashcroft money is buying publicity on a grand scale in these pivotal marginals, the Labour vote is fragmenting. DARLINGTON used to be Tory back in the 1980s, and despite requiring a whopping 13% swing, could go back. This is former Blair acolyte Alan Milburn's seat, and since learning of his numerous outside interests, I'd be sceptical about how hard he's working to keep the seat for his predecessor.
MORLEY & OUTWOOD, where Ed Balls must be worried about being remembered as the 2010 version of Michael Portillo, could be the headline story of the night. The concern in Morley is that the core Labour vote will fragment to the Lib Dems, BNP and UKIP, if it even bothers to turn out, whereas the Tories will get their significant minority out in droves. In a similar scenario, SCUNTHORPE could be vulnerable following the arrest of Labour MP Elliott Morley on expenses charges.
One more Tory bet appeals, on them to win MIDDLESBOROUGH SOUTH and CLEVELAND. Similar arguments apply here about the fragmentation of Labour's core vote, and it looks well within range given that the Tories were supported by over 30% of this electorate even during their darkest days.
20pts CONSERVATIVES TO WIN 300 - 324 SEATS @ 3.5 (VCBET, BLUESQ, BETFAIR)
15pts CONSERVATIVES TO WIN 325 - 349 SEATS @ 4.0 (BETFRED)
15pts CONSERVATIVES TO WIN CREWE & NANTWICH @ 1/2 (PADDY POWER)
15pts CONSERVATIVES TO WIN KEIGHLEY @ 2/5 (PADDY POWER, SKYBET)
6pts CONSERVATIVES TO WIN MORLEY & OUTWOOD @ 2.75 (VCBET, HILLS)
6pts CONSERVATIVES TO WIN MIDDLESBOROUGH SOUTH AND CLEVELAND (2.75 CORALS, VCBET)
3pts CONSERVATIVES TO WIN DARLINGTON @ 5/1 (LADBROKES)
3pts CONSERVATIVES TO WIN SCUNTHORPE @ 5/1 (PADDY POWER, CORALS, HILLS)
BEST LIB DEM & LABOUR BETS
As for Labour and the Lib Dems, I would equally recommend either a straight unders/overs bet or a spread bet. Skybet offer 5/6 for Labour to get lower than 221, well above my 204 prediction, or alternatively we can sell at 206. With the Lib Dems, Skybet also offer 5/6 about them getting over 85, or we can buy at 89. With those spread bets, the upside is far greater than the downside. Even those negative predictions don't factor in the potential for total meltdown. It is quite possible that the remaining core Labour vote either doesn't turn out, or fragments to several smaller parties. There could well be some massive unforeseen shock results out there. That sell of Labour seats also acts as a type of saver against the Tories performing even better than my seats prediction suggests.
Before getting to the plethora of Lib Dem constituency bets, two very small silver linings for Labour supporters. There is one part of the UK where they aren't in freefall - Scotland. Their resilience owes much to a fear of a Tory government, as well as leadership changes and the shine gradually wearing off the SNP administration. In 2005, they lost many votes to the Lib Dems, who were led by a popular Scot in Charles Kennedy, who in contrast to the unpopular Blair, had opposed the Iraq war. Following a brief spell led by another Scot, Ming Campbell, Lib Dem support in Scotland collapsed as they disappeared off the radar and opted for a lesser-known English leader in Clegg. The recent surge has turned that around, but I doubt there will be much difference between the two parties' respective 2005 positions.
If it hadn't been for a freak by-election result in 2006, which saw the Lib Dems pull off a 100/1 upset, DUNFERMLINE and WEST FIFE wouldn't be even on the radar as a vulnerable Labour seat. I expect that they will win it back on a proper General Election turnout, especially given that Gordon Brown may enjoy a rare personal vote as he lives in the constituency. Equally, I don't think they should be outsiders against the Tories in EAST RENFREWSHIRE. There is scant evidence of a Tory revival in Scotland.
BEST BETS ON LABOUR
30pts LABOUR TO GET LESS THAN 220.5 SEATS @ 1.83 (SKYBET)
SELL LABOUR SEATS @ 206
8pts LABOUR TO WIN DUNFERMLINE & WEST FIFE @ 3.0 (SPORTINGBET)
5pts LABOUR TO WIN EAST RENFREWSHIRE @ 2.2 (SPORTINGBET)
BEST BETS ON LIB DEMS
As mentioned above, the key point about the Lib Dems is how their vote is particularly concentrated in the relatively small number of seats they target. They are gradually replacing Labour as the party of the inner-cities, and can expect to pick up seats in Liverpool, Sheffield, Hull, Newcastle and London. University towns are a stronghold, to add to their long-standing tradtions in the South-West and rural Scotland.
Two recent polls are worth mentioning. ICM's marginals poll for the Guardian suggested that the Clegg-bounce was occuring almost entirely within Lab-held targets, rather than the Tory equivalent. And a couple of days earlier, Yougov's regionals poll showed huge swings from Labour to the Lib Dems in the North of England.
The best Lib Dem bets range from odds-on bankers to some tasty outsiders. Big-hitters should have no fears taking the 1.57 about CITY OF DURHAM, especially given its high student population. They need a 4% swing from Labour in this ideal seat, way below the 13% regional swing recorded last week.
Usually in elections, the very biggest names receive a significant personal boost in their seats, and often the surrounding areas. Anyone who watched the debate must surely now know that Nick Clegg is a Sheffield MP, given the number of times he mentioned it, and he can expect to gain SHEFFIELD CENTRAL at the very least. Critically, a significant chunk of the voters in this redrawn constituency were previously in Clegg's Sheffield Hallam, so he would have probably benefitted from a personal vote even without the massive advances of the past fortnight.
My old constituency HULL NORTH is by far their best chance in an Old Labour city that has been turning towards the Lib Dems for years at council level. Again, they should be able to rely on a large university population here, adding to an already impressive regional swing. All three Newcastle seats look within range, the easiest of which should be NEWCASTLE NORTH, while NEWCASTLE-UPON-TYNE CENTRAL is also fair value at 2/1. Again, you may have noticed Clegg referring to this city's Lib Dem council and its return to building council housing, one of the forgotten issues of this election.
In addition to these odds-on bets, the Lib Dems appeal as outsiders in numerous seats. In London, EALING CENTRAL & ACTON, LEYTON & WANSTEAD and WALTHAMSTOW are all plausible gains from Labour in areas where they are increasingly organised. Frank Dobson also looks vulnerable in HOLBORN & ST PANCRAS.
Back in fertile North-East territory, BLYTH VALLEY and REDCAR both look realistic targets. And if there's to be a Clegg-effect in South Yorkshire where he's an MP, PENISTON and STOCKBRIDGE isn't the worst value 6/1 bet out there.
Finally, though the price has been halved since a local poll put them well in front, the Lib Dems could come from fourth place to win Luton South. This is another seat sure to receive media coverage, as Esther Rantzen is running in protest at outgoing Labour MP Margaret Moran's expenses claims. Esther's campaign is apparently struggling, with most defectors of Labour voters
preferring the Lib Dems. Their candidate has a long history in Luton politics, previously for Labour until defecting in 2003. He looks sure to pick up much of the previous Labour vote amongst Luton's Muslim community.
18pts LIB DEMS TO WIN OVER 85.5 SEATS (SKYBET)
BUY LIB DEM SEATS @ 89
20pts LIB DEMS TO WIN CITY OF DURHAM @ 8/15 (HILLS, BET365)
15pts LIB DEMS TO WIN SHEFFIELD CENTRAL @ 10/11 (HILLS)
10pts LIB DEMS TO WIN HULL NORTH @ 1.83 (LADBROKES)
10pts LIB DEMS TO WIN NEWCASTLE NORTH @ 1.8 (VCBET)
6pts LIB DEMS TO WIN REDCAR @ 2.75 (LADBROKES)
5pts LIB DEMS TO WIN HOLBORN & ST PANCRAS @ 3.25 (PADDY POWER, LADBROKES, CORALS)
5pts LIB DEMS TO WIN NEWCASTLE-UPON-TYNE CENTRAL @ 3.0 (PADDY POWER, LADBROKES, CORALS)
4pts LIB DEMS TO WIN BLYTH VALLEY @ 4.0 (LADBROKES, HILLS)
3pts LIB DEMS TO WIN WALTHAMSTOW @ 4.0 (LADBROKES)
3pts LIB DEMS TO WIN LEYTON/WANSTEAD @ 5.5 (CORAL)
6pts LIB DEMS TO WIN EALING CENTRAL/ACTON @ 2.75 (VCBET, PADDY POWER, STAN JAMES, LADBROKES)
3pts LIB DEMS TO WIN LUTON SOUTH @ 6/1 (TOTE, BETFRED, CORALS, SKYBET)
2pts LIB DEMS TO WIN PENISTON & STOCKBRIDGE @ 6/1 (HILLS)
BEST OF THE REST
A few words now about those pivotal Tory/Lib Dem marginals. As stated above, I've estimated no net change overall between them, although several will change hands.
Lib Dem MPs are generally very good at building a personal vote in their constituencies, and are hard to replace. Still up to a dozen seats are vulnerable to a resurgent Tory party, most notably; St Austell & Newquay, Carshalton & Wallington, Sutton & Cheam, Torbay and Herefordshire South. In return, they have a good chance of picking up Tory seats such as Eastbourne, St Albans, Wells, Filton & Bradley Stoke.
Boundary changes also confuse matters slightly, altering the effects of incumbency. So on the one hand, the Lib Dems could struggle to retain
Chippenham, which they hold on notional figures, but where they have no sitting MP. Alternatively,
Solihull has a Lib Dem MP, yet is notionally Conservative on the adjusted figures. My prediction is for Lorely Burt to hold off the Tory challenge thanks to the benefit of incumbency.
Finally, of all the various independents challenging in this election, BOB SPINK could be the one to grab the headlines. Spink has been elected three times as Tory MP for Castle Point, winning the seat back from Labour in 2001. He was expelled from the party a couple of years ago after a series of spats, including a row over anti-immigration leaflets.
Spink is running a well-financed campaign against a young, centrally imposed candidate. I suspect his brand of right-wing Euroscepticism is closer to the mood of this particular electorate than Cameron's Conservatism, and will ensure he retains a substantial chunk of previous supporters. Moreover, given a long record of championing local issues, now he's shorn of the Tory tag, Spink may be able to attract decent numbers from disillusioned Labour voters.
ADVISED BET
4pts BOB SPINK TO WIN CASTLE POINT @ 3/1 (GENERAL)
